1. e21网站验证码识别
"""__author__= 雍新有"""
from io import BytesIO
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.wait import WebDriverWait
from PIL import Image
from chaojiying_Python.chaojiying import main1
browser = webdriver.Chrome()
browser.get('http://bm.e21cn.com/login/user')
wait = WebDriverWait(browser, 10)
# 将屏幕的宽高自定义,或者执行js实现拖拽(window.scrollTo(1000, 1000))
# browser.set_window_size(1500, 1300)
def screen_big_png():
# 获取整个窗口的图片
big_screen = browser.get_screenshot_as_png()
# 保存 BytesIO -- 读取二进制文件
img = Image.open(BytesIO(big_screen))
print(img)
img.save('a1.png')
return img
def get_position():
# 显示等待
img = wait.until(
EC.presence_of_element_located((By.XPATH, '//*[@id="imgCheckCode"]'))
)
print(img.location)
print(img.size)
size = img.size
location = img.location
# 左上角定位
x1 = location['x'] * 1.25
y1 = location['y'] * 1.25
# 右下角定位
x2 = x1 + size['width']*1.28
y2 = y1 + size['height']*1.28
return (x1, y1, x2, y2)
def screen_small_png(big_png):
# 先获取验证码的位置,x和y
x1, y1, x2, y2 = get_position()
img = big_png.crop((x1, y1, x2, y2))
img.save('a2.png')
if __name__ == '__main__':
# 扣大图
big_png = screen_big_png()
# 扣小图
screen_small_png(big_png)
# 超级鹰校验
result = main1('a2.png')
code = result['pic_str']
print(code)
# 模拟登陆
# 显示等待,获取
2. B站极验验证码
"""__author__= 雍新有"""
import time
from io import BytesIO
from selenium import webdriver
from selenium.webdriver import ActionChains
from selenium.webdriver.support.wait import WebDriverWait
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as EC
from PIL import Image
class BiliSpider():
def __init__(self):
self.browser = webdriver.Chrome()
self.wait = WebDriverWait(self.browser, 30)
self.url = 'https://passport.bilibili.com/login'
self.username = 'coco'
self.password = '123456'
self.filename1 = 'big1.png'
self.filename2 = 'big2.png'
self.smallname1 = 's1.png'
self.smallname2 = 's2.png'
# def __del__(self):
# # 类执行完后会自动调用这个函数
# self.browser.close()
def login_open(self):
# 打开B站登陆页面,并输入账号密码,最后点击登陆按钮
self.browser.get(self.url)
# 账号输入框
name_input = self.wait.until(
EC.presence_of_element_located((By.XPATH, '//*[@id="login-username"]'))
)
name_input.clear()
name_input.send_keys(self.username)
# 密码输入框
password_input = self.wait.until(
EC.presence_of_element_located((By.XPATH, '//*[@id="login-passwd"]'))
)
password_input.clear()
password_input.send_keys(self.password)
# 点击登陆按钮
button = self.wait.until(
EC.element_to_be_clickable((By.XPATH, '//*[@id="geetest-wrap"]/ul/li[5]/a[1]'))
)
button.click()
# 目的是让验证码加载出来,手动的睡眠几秒
time.sleep(3)
def save_big_png(self, filename, smallname):
# 实现截大图
img = self.browser.get_screenshot_as_png()
img = Image.open(BytesIO(img))
img.save(filename)
# 截取小图
small_png = self.crop_png(img, smallname)
return small_png
def screen_png(self):
# 截大图
# 横向滚动
# js = 'window.scrollTo(1000, 0)'
# self.browser.execute_script(js)
# 截取,保存有缺口的大图, 返回小图
img_s1 = self.save_big_png(self.filename1, self.smallname1)
# 隐藏验证码中的缺口,然后在截取
js = 'document.getElementsByClassName("geetest_canvas_fullbg")[0].style.display="block"'
self.browser.execute_script(js)
img_s2 = self.save_big_png(self.filename2, self.smallname2)
return img_s1, img_s2
def get_position(self):
# 获取左上角和右下角的横纵坐标位置
chapter = self.wait.until(
EC.presence_of_element_located((By.XPATH, '/html/body/div[2]/div[2]/div[6]/div/div[1]/div[1]/div/a/div[1]/div/canvas[2]'))
)
location = chapter.location
size = chapter.size
x1 = location['x']
y1 = location['y']
x2 = x1 + size['width']
y2 = y1 + size['height']
return x1, y1, x2, y2
def crop_png(self, img, filename):
# 截取小图,有缺口小图和无缺口小图
x1, y1, x2, y2 = self.get_position()
small_img = img.crop((x1, y1, x2, y2))
small_img.save(filename)
return small_img
def compare_img(self, img1, img2, x, y):
# 比较图片像素点,像素点相似返回True,否则False
# getpixel((x, y)) , img1.load()[x, y] - 获取图片像素点的rgba值
pix1 = img1.load()[x, y]
pix2 = img2.load()[x, y]
# 阈值 - 像素偏差
a = 60
if abs(pix1[0] - pix2[0]) < a and \
abs(pix1[1] - pix2[1]) < a and \
abs(pix1[2] - pix2[2]) < a and \
abs(pix1[3] - pix2[3]) < a:
# 两个像素点相差不大
return True
return False
def get_distance(self, img1, img2):
# 计算两张小图的缺口距离
# 比较两张图片的每一个像素点,误差不能超过某个阈值
print(img1.size)
left = 70
# 遍历小图中横坐标58右边的所有点
for x in range(left, img1.size[0]):
for y in range(img1.size[1]):
# 比较2张小图的像素点
if not self.compare_img(img1, img2, x, y):
return x
return left
def slider_button(self, distance):
# 拖动下面滑块
slider = self.wait.until(
EC.presence_of_element_located((By.XPATH, '/html/body/div[2]/div[2]/div[6]/div/div[1]/div[2]/div[2]'))
)
action = ActionChains(self.browser)
# 执行点击并抓住
action.click_and_hold(slider).perform()
print('==============')
while distance > 0:
print(distance)
distance -= 2
action.move_by_offset(xoffset=2, yoffset=0).perform()
# 新建ActionChains对象防止累加位移
action = ActionChains(self.browser)
# time.sleep(0.2)
action.release(slider).perform()
def start(self):
self.login_open()
img_s1, img_s2 = self.screen_png()
# 获取两张小图的距离 -- 从图中滑块的左边到阴影的左边
distance = self.get_distance(img_s1, img_s2) - 7
print(distance)
# 滑动滑块
self.slider_button(distance)
if __name__ == '__main__':
# 扣有缺口图和没有缺口图,对比两张图的像素点,找出拖拽的横坐标,实现拖拽。
spider = BiliSpider()
spider.start()